Nozzles sizing for separator

[kald]

in design of two phase and three phase separator , what the factors must be consider while outlet Nozzles sizing for the separator , when i size using some excel spreadsheet the nozzle can't easily handle the liquid level in the separator and when increase this size it handle out this liquid quickly and protect the separator from high level in abnormal operation situation (ex.slug flow). so is any constraint in increasing outlet nozzle size

 

[chengg29]

Generally the equipment nozzles are sized to match the connected pipe work and the connected pipe work are sized to meet pressure drop and velocity requirements based on a a given flow of fluids through them (some design standards like the Shell DEP on 2 phase / 3 phase Separators mention a maximum velocity criteria of 1 m/s to be maintained in the liquid outlet nozzles of a separator. However Shell appears to mention this criteria more to typically align with an industry practice of maintaining low velocities in typical pump suction lines connected to vessels).

A non-pump suction liquid line could be sized using a general velocity criteria of 3 m/s and an appropriate pressure drop (assuming the liquid is clean as most hydrocarbon liquids are. Exceptions could be hydrocarbon slurry like liquids with a propensity to deposit solids at low pipe velocities or high viscosity liquids and / or any other special application liquids). That said however each company may have their own set criteria for maximum recommnded velocities to be maintained in the vessel nozzles and I reckon your company may have one too which I think you must follow.

Usually pressure drops for typical in-plant liquid pipe work of single phase of few odd hundred meters length should not be of a concern until and unless its otherwise. However it will be different if the flow has some semblance of two phase flow (slug flow for example) and sizing for two phase flow is not as straight forward as single phase. However calculation techniques / methods exist as also specialised software / spreadsheets do exist. In most situations the flow regime (slug, annular etc.) as determined by calculations on paper may not eventually be the same when the actual facility is installed and operated. the two pahse flow may end up in a totally different flow regime.

I assume the spreadsheet that you mention is the company's validated official spreadsheet for doing such calculations. Hence it would be prudent first re-visit your calculations on it.

Cannot say more than this since you dont seem to mention any details of your problem.

cheers

 

[MortenA]

@kald, i dont know if you used google translate but i have a hard time following your question (maybe its just me, im not from an English speaking country).

However, i would give you one piece of advice: Choose a nozzle size one large than required (and a reducer downstream) and make both water and oil nozzle the same size (the largest).

Your mechanical engineer will try to get smaller nozzles because they cost money - but having to replace its with a larger nozzle at a later stage is many times more costly!

Best regards, Morten

 

[kald]

very thanks chengg29 for your feedback and valuable information
actually MortenA your advice what i ask for that i can increase nozzle size as i can without any constraints except of cost , because that i can solve some operation problem in the future

with reference to english language , i hope that if any one have the same issue with my language notify me , to try improve it . is my language hard to be understood , is there any problem in formation of english sentence

 

[dingyong]

@kald ,have a constraint with increasing the size of nozzle ,the velocity is 0.9~1.5m/s for liquid outlet

 

[gl99]

@Kald,
Normally, you can size the inlet nozzle, gas outlet nozzle and liquid outlet nozzle based on the below critical,
Inlet nozzle: rho(mix) * v2 < 6000 lb/ft.sec2
Gas outlet nozzle: rho * v2 < 3000 lb/ft.sec2
Liquid Outlet: velocity < 3.3 ft/sec
You can find lots detail engineer tools in www.enground.com.